Skip to content
Aakash GuptaAakash Gupta

I can’t believe we built an AI employee in 62 mins (Cursor, ChatGPT, Gibson)

This is another episode from our AI PM series. This time, we’re building an AI teammate that runs user research, writes product docs, and powers customer success end-to-end with GibsonAI founder, Harish Mukhami. We're building: Preview – 00:00:00 Building AI Customer Success Agent (Tool Stack) – 00:01:46 Role of GibsonAI in Building Customer Success AI Agent – 00:07:29 Using Data from O3 Mini – 00:09:20 Ad (Amplitude) – 00:10:13 Ad (Linear) – 00:10:45 Directing GibsonAI – 00:11:45 Connecting GibsonAI via MCP – 00:17:38 Role of Cursor – 00:21:10 Python Script Inserting Data – 00:26:56 Understanding Cursor Modes – 00:29:00 Ad (Maven) – 00:30:38 Our Dashboard Is Ready – 00:31:01 Building AI Agent – 00:33:44 The the Most Important Thing Our Agent Is Doing – 00:41:46 Aakash’s Reaction to Output – 00:50:51 Role of CrewAI – 00:52:01 AI Employee Use Cases for PMs – 00:54:47 Why Harish Built GibsonAI – 00:56:35 Final Thoughts – 01:00:15 Podcast transcript: https://www.news.aakashg.com/p/harish-mukhami-podcast 💼 Check out our sponsors: Amplitude: The market-leader in product analytics - https://amplitude.com/session-replay?utm_campaign=session-replay-launch-2025&utm_source=linkedin&utm_medium=organic-social&utm_content=productgrowthpodcast Linear: Plan and build products like the best - https://linear.app/partners/aakash Maven: Check out my own curation of their courses for a discount - http://maven.com/x/aakash 👀 Where to Find Harish LinkedIn:https://www.linkedin.com/in/harishmukhami GibsonAI: https://www.gibsonai.com/?utm_medium=podcast&utm_source=aakash 👨‍💻 Where to find Aakash: Twitter: https://www.twitter.com/aakashg0 LinkedIn: https://www.linkedin.com/in/aagupta/ Instagram: https://www.instagram.com/aakashg0/ 🔑 Key Takeaways 1. Production Over Prototypes - Stop building prototypes and start shipping production-ready AI employees. Gibson AI, Cursor, and CrewAI let you go from concept to production in hours. Harish's agent was backed by a scalable database handling 10,000 users day one—no rebuilding required. 2. Amplify, Don't Replace - Your next 10x gain comes from making existing teams superhuman. AI agents analyze dashboards 24/7 and draft personalized outreach, while human CS agents focus on high-touch relationships and strategic decisions. 3. Three-Tier Implementation Strategy - Follow this roadmap: dashboard → human-approved recommendations → autonomous actions. Start with AI insights humans review, then AI recommendations humans approve, finally autonomous execution for low-risk tasks. 4. Human-Loop Insurance - Human-in-the-loop is customer relationship insurance. Harish built approval workflows because random AI emails "will only make the problem worse." AI should amplify human judgment, not bypass it. 5. Proactive Beats Reactive - Proactive churn prevention beats reactive win-back by orders of magnitude. AI agents monitor engagement patterns and usage metrics to address churn risks before customers consider leaving. 6. MCP Integration Magic - MCP makes AI tools actually talk to each other. Harish could query databases, update schemas, and deploy changes directly from Cursor—seamless integration without manual tool switching. 7. Information Processing Automation - Any role that "ingests information and sends out information" is automatable. SDRs, recruiters, executive assistants—if it involves processing data and taking action, AI handles the heavy lifting. 8. Specialized Model Selection - Different models excel at different tasks. Harish used O3 Mini for planning, Claude Sonnet for coding. Match your model choice to the specific job rather than defaulting to popularity. 9. Day-One Infrastructure - Production-grade infrastructure eliminates the prototype-to-production death valley. Starting with scalable database infrastructure means your demo can actually handle real user volumes when stakeholders want to scale. 10. Always Review Code - Read AI-generated code even when moving fast. Despite impressive capabilities, human oversight remains critical: "Make sure it is the code that you want." Speed matters, but understanding what you ship is non-negotiable. #ai #aiagents #agents 🧠 About Product Growth: The world's largest podcast focused solely on product + growth, with over 170K listeners. Hosted by Aakash Gupta, who spent 16 years in PM, rising to VP of product, this 2x/ week show covers product and growth topics in depth. 🔔 Subscribe and like the video to support our content! And turn on the bell for notifications.

Harish MukhamiguestAakash Guptahost
May 31, 20251h 2mWatch on YouTube ↗

At a glance

WHAT IT’S REALLY ABOUT

Build an AI customer success employee using Cursor, Gibson, CrewAI fast

  1. The build is structured into three phases: a SaaS-style analytics dashboard, a human-in-the-loop agent that recommends actions, and a fully autonomous agent that executes workflows.
  2. OpenAI o3-mini is used for upfront planning of dashboard tabs/metrics and required data fields, while Cursor (Claude Sonnet 3.7) generates the application code and scripts.
  3. Gibson acts as an AI-managed cloud database that generates a data model, deploys dev/prod environments, and exposes instant CRUD APIs that the dashboard and agents consume.
  4. CrewAI orchestrates multiple specialized agents (query, churn analysis, mitigation, ingestion) and later adds action agents to send SendGrid emails and create Jira tickets automatically.
  5. The conversation emphasizes practical infrastructure realities—schema evolution, scaling, API-driven data access, evals, and the need to review AI-generated code before deploying autonomous behavior.

IDEAS WORTH REMEMBERING

5 ideas

Start with clear dashboard goals before writing any code.

They prompt o3-mini as a BI expert to define four tabs (funnel, engagement, health risk, retention/churn) and then derive the needed data fields, preventing ad-hoc schema decisions later.

Model your database around real data sources, not just UI screens.

After the tab layout, they re-organize requirements by source (HubSpot/CRM, Google Analytics, app/transactional DB), which maps better to ingestion pipelines and schema design.

AI-managed databases can compress backend setup from days to minutes.

Gibson generates the schema/ERD, deploys to cloud with dev/prod environments, and exposes a CRUD API layer quickly—removing much of the traditional setup, scaling, and migration burden.

MCP turns the IDE into an operations console for your backend.

By connecting Gibson to Cursor via MCP, they manage projects, fetch schemas, redeploy changes, run natural-language queries, and keep developers inside the coding environment.

Use scripted test data to make demos repeatable and realistic.

Cursor generates a Python + Faker-based loader that respects table relationships and uses Gibson APIs, ensuring the dashboard renders believable charts without hand-entering rows.

WORDS WORTH SAVING

5 quotes

Today we'll be building an AI customer success agent, uh, which is gonna solve problems for you without any humans involved.

— Harish Mukhami

One of my favorite use cases with Gibson, Aakash, that you have just seen, is you plug Gibson into vibe coding tools, um, um, and you're not building a prototype anymore.

— Harish Mukhami

It's already deployed in the cloud, so you publish this and you get 10,000 users tomorrow, we can handle that.

— Harish Mukhami

The barrier to build, uh, is so, so low that I feel like h-how did we not have this two, three years ago? I wish I had this few years ago, so.

— Harish Mukhami

Any code that AI generates, please read. Uh, make sure that you don't just blindly accept.

— Harish Mukhami

Three-phase AI employee build (dashboard → recommendations → autonomy)Dashboard design via LLM planning (tabs, KPIs, data requirements)Gibson AI database: schema generation, ER diagram, deploy, CRUD APIsMCP integration connecting IDE to database operationsCursor agent mode for Next.js dashboard + test data generationCrewAI multi-agent orchestration (roles, tasks, tools)Autonomous actions: SendGrid email + Jira ticket creation; human-in-loop + evals

High quality AI-generated summary created from speaker-labeled transcript.

Get more out of YouTube videos.

High quality summaries for YouTube videos. Accurate transcripts to search & find moments. Powered by ChatGPT & Claude AI.

Add to Chrome